Machine learning in prediction of stock market indicators based on historical data and data from Twitter sentiment analysis.
Development of linguistic technologies and penetration of social media provide powerful possibilities to investigate users’ moods and psychological states of people. In this paper we discussed possibility to improve accuracy of stock market indicators predictions by using data about psychological states of Twitter users. For analysis of psychological states we used lexicon-based approach, which allow us to evaluate presence of eight basic emotions in more than 755 million tweets. The application of Support Vectors Machine and Neural Networks algorithms to predict DJIA and S&P500 indicators are discussed.
In work the developed model of adaptive management by the vertically integrated companies based on the system approach supporting the mechanism of an operational management in a uniform cycle of strategic planning, within the limits of faster time is presented. Thus for a finding of optimum values of operating parameters special algorithms of a class of genetic algorithms are used, neural networks the example of the developed system of adaptive management for the vertically-integrated oil company is etc. presented.
The article discusses development of the segmented characters classifier of the Russian alphabet a nd of the Arabic numerals on the basis of block neural network structures including the plurality of blocks for each individual character recognition and for the synthesis block decision. Keywords: pattern recognition, neural network, training of neural n etworks, base of hand - written characters, recognition of hand - written characters
This book constitutes the refereed proceedings of the 6th IAPR TC3 International Workshop on Artificial Neural Networks in Pattern Recognition, ANNPR 2014, held in Montreal, QC, Canada, in October 2014. The 24 revised full papers presented were carefully reviewed and selected from 37 submissions for inclusion in this volume. They cover a large range of topics in the field of learning algorithms and architectures and discussing the latest research, results, and ideas in these areas.
The question about possibilities to use Twitter users’ moods to increase accuracy of stock price movement prediction draws attention of many researchers. In this paper we examine the possibility of analyzing Twitter users’ mood to improve accuracy of predictions for Gold and Silver stock market prices. We used a lexicon-based approach to categorize the mood of users expressed in Twitter posts and to analyze 755 million tweets downloaded from February 13, 2013 to September 29, 2013. As forecasting technique, we select Support Vector Machines (SVM), which have shown the best performance. Results of SVM application to prediction the stock market prices for Gold and Silver are discussed.
The Caucasus is the place with the greatest linguistic variation in Europe. The present volume explores this variation within the tense, aspect, mood, and evidentiality systems in the languages of the North-East Caucasian (or Nakh-Daghestanian) family. The papers of the volume cover the most challenging and typologically interesting features such as aspect and the complicated interaction of aspectual oppositions expressed by stem allomorphy and inflectional paradigms, grammaticalized evidentiality and mirativity, and the semantics of rare verbal categories such as the deliberative (‘May I go?’), the noncurative (‘Let him go, I don’t care’), different types of habituals (gnomic, qualitative, non-generic), and perfective tenses (aorist, perfect, resultative). The book offers an overview of these features in order to gain a broader picture of the verbal semantics covering the whole North-East Caucasian family. At the same time it provides in-depth studies of the most fascinating phenomena.
The paper theorizes on the general architectonics of idealized cognitive models (ICMs) and their involvement in metonymy and metaphor. The article posits that an ICM's structure should reflect the architecture of the neural network/s engaged in processing of a given concept. The ICM nodes, or cogs, construct a complex, hierarchically organized neural connections, with the superior nodes being highly selective, invariant and prototypical. Signals travelling from one cog to another within one ICM are essentially metonymical, while a cog shared by two or more ICMs marks a metaphoric shift.
The form whose main function is to express indirect commands, called the third person Imperative, Jussive or Exhortative, when compared to the prototypical (second person) Imperative, shows semantic and formal similarities and distinctions at the same time. The study describes formal and functional patterns of Jussive and places this category within the typology of the related categories, such as Imperative and Optative, based on data from six East Caucasian languages (Archi, Agul, Akhvakh, Chechen, Icari and Kumyk). Five formal patterns of Jussive are attested in these languages, including a specialized form, constructions derived from want, from tell him to do and from make him do and the Optative. Jussive forms may express such meanings as third person command, indirect causation, permission, indifference towards the accomplishment of an action and an assumption. While the Jussive is crucially different from the second person Imperative in that it introduces a third participant, this article shows that it is the addressee, not a third person, who is the central participant of a Jussive situation from both formal and functional points of view.